Path: csiph.com!eternal-september.org!feeder.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Ben Bacarisse <ben.usenet@bsb.me.uk>
Newsgroups: comp.lang.c
Subject: Re: Is this valid C?
Date: Fri, 03 Mar 2017 12:57:22 +0000
Organization: A noiseless patient Spider
Lines: 39
Message-ID: <87fuiuldql.fsf@bsb.me.uk>
References: <o8858o$g1g$1@dont-email.me> <o91ap7$q14$1@jstuckle.eternal-september.org> <o91b7n$t1n$1@dont-email.me> <o91hio$lsd$1@jstuckle.eternal-september.org> <o91kds$186$1@dont-email.me> <o9216d$k7o$1@jstuckle.eternal-september.org> <o9246f$n5$1@dont-email.me> <o9283r$eav$1@jstuckle.eternal-september.org> <o92c4n$vt9$1@dont-email.me> <o92dil$4h6$1@jstuckle.eternal-september.org> <o92epd$8j3$1@dont-email.me> <o92h8g$eo8$1@jstuckle.eternal-september.org> <o92nd3$tst$1@dont-email.me> <o92u2s$7a0$1@jstuckle.eternal-september.org> <o930jl$cmi$1@dont-email.me> <o94bpj$pm5$1@jstuckle.eternal-september.org> <97ad8e6e-6ed3-474a-9c77-dbf3a11e3c93@googlegroups.com> <o9a0nd$snc$1@dont-email.me> <o9bg9h$rga$1@dont-email.me> <87lgsmleh9.fsf@bsb.me.uk> <o9bohv$nrn$1@dont-email.me>
Mime-Version: 1.0
Content-Type: text/plain
Injection-Info: mx02.eternal-september.org; posting-host="daac0805028715f3f2958ec41815e0d4"; logging-data="23319"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX182V7EcFTl2EbQxPbhBovI5gmA+PJsyvj4="
Cancel-Lock: sha1:DI9H9dxcl1FzCwzeOE1owen9oHU= sha1:FPU48pYO+D5Z6vwOsmIuKWSrONU=
X-BSB-Auth: 1.1e09ccc3c49fde95eb0d.20170303125722GMT.87fuiuldql.fsf@bsb.me.uk
Xref: csiph.com comp.lang.c:105350

Richard Heathfield <rjh@cpax.org.uk> writes:

> On 03/03/17 12:41, Ben Bacarisse wrote:
>> David Brown <david.brown@hesbynett.no> writes:
>>
>>> On 02/03/17 21:54, James Kuyper wrote:
>>>> Jerry's uncharacteristic silence has given me some time to think about
>>>> this issue more carefully. The idea that operator precedence and
>>>> associativity always determine the order of evaluation of C expressions,
>>>> with expressions with a high precedence always being
>>>> evaluated first, and associativity being used to break ties between
>>>> expressions with the same precedence is very popular, and has been
>>>> around for a very long time. Jerry didn't invent it, not by a long
>>>> shot, but I need a simple name for this rule, so I'll call it the
>>>> "Stuckle rule" in honor of Jerry Stuckle.
>>>>
>>>
>>> (I snipped the rest of your detailed and informative post, simply to
>>> save space.)
>>>
>>> Would it be fair to say that if you have an expression where no
>>> sub-expression has any side effects (no volatile accesses, ++ operators,
>>> assignment operators, etc., and only "pure" function calls) then the
>>> Stuckle rule will apply?  If you have "x && y" and "y" has no side
>>> effects, then it makes no difference to the outcome of the program
>>> whether y (or its sub-expressions) is evaluated, even if x evaluates
>>> to 0.
>>
>> It depends on what you mean by saying that the rule will apply.
>
> Well, it never *actually* applies. But there are circumstances in
> which the implementation may behave "as if" it applies.

And, curiously, you cut the very next sentence where I said exactly
that. :-(

<snip>
-- 
Ben.